The purpose of the study presented in this paper is to compare the current ACI 318-11 strut-and-tie modeling procedure to the previous ACI 318-99 sectional approach for the design of deep beams. To accomplish this goal, the author compiled a database of 905 deep beam tests from the literature. Of these tests, 438 specimens were isolated based on available specimen details, configuration of the test, or both. The shear capacity of these specimens was calculated according to the ACI 318-99 (Section 11.8) and ACI 318-11 (Appendix A) provisions, and compared with their experimentally measured capacity. Based on this analysis, it was observed that capacities calculated with both provisions are statistically similar for the experimental data available; however, the average of the capacities calculated per the ACI 318-99 procedure is increasingly unconservative as the specimen's cross-sectional area increases. The implication of this finding is that use of the ACI 318-99 provisions to design a conventionally sized transfer girder may potentially result in an unsafe situation. The strut and tie modeling procedure did not exhibit this downward trend because the method is founded in first principles. Thus, structural engineers can be assured that their transfer girders will be conservatively designed when using the strut and tie modeling procedure.